video-gen

Generate AI video clips from text, images, and reference footage.

751|69|Updated Jun 25, 2026
One-click install
npx skills add https://github.com/ChatCut-Inc/agent-plugin --skill video-gen-chatcut-inc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: video-gen
Source: https://github.com/ChatCut-Inc/agent-plugin/tree/main/claude/skills/video-gen
Command: npx skills add https://github.com/ChatCut-Inc/agent-plugin --skill video-gen-chatcut-inc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps users turn creative ideas, images, and existing footage into coherent AI-generated video clips without guessing model settings, reference inputs, or revision workflows.

Core Features & Use Cases

  • Multi-Model Video Generation: Create text-to-video, image-to-video, first/last-frame transitions, and reference-guided clips with Seedance 2.0, Seedance 2.0 mini, Kling, or Gemini Omni.
  • Editing and Iteration: Modify existing clips, extend or bridge footage, preserve visual consistency across shots, and select the appropriate model for localized edits, drafts, or high-quality generation.
  • Production Guardrails: Align duration, shot structure, content, and consistency anchors before submission; validate model-specific parameters; use project asset references; and hand job tracking to the progress workflow.

Quick Start

Use the video-gen skill to create a cinematic 9:16 video of a cyclist riding through a neon-lit city at night for eight seconds.

Frequently Asked Questions about video-gen

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate AI video clips from text prompts or images?

To generate AI video clips, you provide text prompts, images, or reference footage. The Skill validates model-specific parameters and project asset references to create text-to-video or image-to-video outputs using Seedance, Kling, or Gemini Omni.

Can I extend existing footage or create first and last frame transitions?

Yes, you can extend or bridge existing generated clips and create first and last-frame transitions. The Skill preserves visual consistency across shots and applies localized edits by selecting the appropriate AI video generation model.

What is the best way to maintain visual consistency across multiple storyboard shots?

Maintaining visual consistency across multi-shot storyboards requires using consistency anchors and project asset references. The Skill aligns duration, shot structure, and content before submission to ensure coherent AI video generation.

Does this video generation workflow support Kling and Gemini Omni models?

Yes, the workflow supports multi-model video generation using Seedance 2.0, Seedance 2.0 mini, Kling, and Gemini Omni. It validates model-specific parameters to ensure high-quality text-to-video and reference-guided clip output.

How do I track the progress of my video generation jobs?

You track video generation jobs through the track_progress workflow. After submitting validated prompts and model-appropriate parameters, the system hands job tracking to this workflow to monitor the AI video rendering process.

Why do I need to use descriptive asset names and project asset references for AI video?

Descriptive asset names and project asset references are required to align duration, shot structure, and consistency anchors. These production guardrails prevent guessing model settings and ensure the AI video generation matches your creative intent.